EP3582227A1 - Image recording device, computer program, electronically readable data carrier and method for operating a medical image recording device - Google Patents

Abstract

Method for operating a medical image recording device (8) in the context of an image recording process on a patient (14) serving an image recording target, the image recording device (8) being controlled on the basis of operating parameters to be implemented by a control device (15) of the image recording device (8), wherein the operating parameters by means of a determination algorithm (6) completely automatically from at least the patient (14) in the form of a patient model and / or input data (2, 3, 4, 5) describing the image acquisition target and a registration (1) of the patient (14) with a Coordinate system of the image recording device (8) is determined and used to control the image recording device (8).

Description

The invention relates to a method for operating a medical image recording device as part of an image recording process on a patient serving an image recording target, the control of the image recording device taking place on the basis of operating parameters to be implemented by a control device of the image recording device. In addition, the invention relates to a medical image recording device, a computer program and an electronically readable data carrier.

Modern medical image acquisition devices represent complex systems, the operation of which presents users with great challenges in order to achieve a suitable or even optimal image quality during an image acquisition process on a patient. It is therefore known in the prior art to offer users manually selectable, predefined measurement protocols to choose from, so that they can select a suitable measurement protocol based on the image acquisition target of the current image acquisition process. Such measurement protocols are also referred to as “organ programs”, particularly in the case of X-ray devices as image recording devices.

The measurement protocols, which define operating parameters to be implemented by a control device of the image recording device, are extremely different from the possible image recording goals and medical disciplines. For example, measurement protocols for both different diagnostic examinations and interventions in the field of cardangiography, electrophysiology, neurointervention, general angiography, pediatrics, surgery, etc., can be offered on different organs, for example the heart, liver, head, legs, spinal area, etc. , The measurement reports differ in their operating parameters, for example physical operating parameters, such as dose, tube voltage, pulse length, image frequency, detector resolution, detector mode, tube focus selection and the like in X-ray devices, and / or operating parameters related to the evaluation, such as native mode, subtraction mode, general image superimposition (for example fluoroscopy and 3D Image data set), type of image-based or temporal noise suppression, frequency-dependent contrast enhancement and / or contrast reduction, management of image dynamics, image sharpness and the like. Overall, it can therefore be said that there is an increasing number of optimized measurement protocols on an image recording device with an increasing number of possible uses and areas of application, which provide suitable operating parameters, be they image recording parameters or image evaluation parameters.

For a respective image acquisition process, which can also form an examination section and / or treatment section, for example a diagnostic portion of a treatment process, placement of medical instruments such as a balloon, stent, coil, catheter for ablation of embolization materials or for the administration of chemotherapeutic materials, biopsy needles and the like , the user, for example a doctor who carries out the diagnosis / therapy, selects an appropriate measurement protocol which he believes is suitable.

In daily practice it can often happen that the optimal measurement protocol for the image acquisition process, in particular organ program, is not used after, for example, the doctor as the user puts his focus and concentration on the examination / therapy and the patient or after training, in spite of it is not aware that an alternative measurement protocol might deliver better image acquisition results. In order to improve this, it has already been proposed to combine measurement protocols into a sequence for an overall examination or an overall treatment and thus to provide a defined sequence of measurement protocols for different examination or treatment sections. However, such a summary of measurement protocols does not solve the problem of the large number of measurement protocols that are available manually.

The invention is therefore based on the object of specifying a possibility for controlling the operation of an image recording device which simplifies the operation of an image recording device and contributes to improving the image recording quality.

This object is achieved by a method with the features of claim 1, an image recording device with the features of claim 15, a computer program with the features of claim 16 and an electronically readable data carrier with the features of claim 17. Advantageous embodiments result from the subclaims ,

A method according to the invention for operating a medical image recording device provides that the operating parameters are determined completely automatically by means of a determination algorithm from at least the patient in the form of a patient model and / or input data describing the image recording target and a registration of the patient with a coordinate system of the image recording device and for controlling the image recording device be used.

A radical departure from the known state of the art, which provides various measurement protocols that are selected manually, is aimed at by no longer defining any measurement protocols in the image recording device that are selected either purely manually or supported by the system, but instead using an inherent system Intelligence according to the determination algorithm has a direct influence on the physical and image processing operating parameters that are relevant for the respective image acquisition process, in particular as an examination and / or treatment section are optimal. In other words, the invention proposes that an appropriately enabled determination algorithm takes over large parts or even the complete image-relevant parameterization and thus minimizes or completely excludes the user interaction, for example in the form of a measurement protocol selection.

Such a determination algorithm, which, by the way, cannot necessarily be chosen comprehensively, but in a preferred exemplary embodiment can use artificial intelligence extensively, uses in particular input data describing the patient in the form of a patient model and an existing registration of the patient with a coordinate system of the image recording device, which means on the basis of The position of the patient model in relation to the image recording device is also known from the registration, so that it can in particular be concluded which examination area is to be recorded, which in turn gives clear indications of the image recording target. In particular with regard to the image acquisition target, further input data can also be present, as will be explained below, in order to be able to conclude together as precisely as possible which operating parameters deliver the best imaging result for the type of imaging of the image acquisition device. Generally speaking, further input parameters can relate to system and environmental conditions, in particular medical instruments and / or aids used for the image recording process. Taking into account the position of the patient and thus his anatomical structures relative to the image recording device or the corresponding recording arrangement of the image recording device, an information base that can be continuously updated is also created, which can suitably track operating parameters in dynamically changing examination areas, in particular when tracking a medical instrument, for example one Catheter, through the body of a patient, which is to be guided, for example, from the arm or from the groin to the target location, with these different entry points clearly different and suitable operating parameters suggest that can be determined automatically on the basis of the input data and the existing registration by the determination algorithm and can be used in the control device.

In addition to image acquisition parameters (physical operating parameters), the operating parameters can also include at least one image evaluation parameter (image processing parameters) with particular advantage. After many (pre-) evaluation steps with regard to the image recording result, that is to say the image data, have already been carried out within the image recording device, it is therefore particularly expedient to also adapt the image processing with regard to the image recording target described by the input data and thus at least derivable from it, that an improved image quality also arises in this regard, without corresponding manual inputs by a user being required. It is particularly advantageous that the operating parameters that can be determined automatically by means of the determination algorithm therefore include all operating parameters that have been provided or can be provided so far by measurement protocols, so that the determination algorithm can automatically put together a measurement program that is individually adapted to the current image acquisition process and the current patient based on the input data.

The determination algorithm therefore uses decision criteria which evaluate the input data or intermediate data derived therefrom at least partially explicitly or at least partially implicitly, for example by means of artificial intelligence. An essential decision criterion for the present invention is the use of the information given by the input data to determine the organ (OOI - Organ of Interest) located in the field of view of the image capturing device, in order to automatically select the optimal physical and, in particular, image processing technology operating parameters.

An expedient development of the present invention provides that the determination algorithm comprises at least one artificial intelligence algorithm that was trained by machine learning, in particular by means of deep learning. Artificial intelligence has proven to be a tool that is not absolutely necessary, but extremely useful and improves the overall quality of the automatically determined operating parameters in the context of the present invention. Especially in connection with the increasing complexity of image recording devices, in particular due to the increasing equipment and / or increasing number of setting options, the artificial intelligence lends itself to extract relationships between input parameters and operating parameters and, if appropriate, also operating parameters and operating parameters and after a corresponding learning process for further improvement the quality of the image acquisition result. Suitable training data are used for this.

In this context, a particularly expedient embodiment of the present invention provides that training data for the artificial intelligence algorithm are determined by logging an activity, in particular comprising at least one individual parameter adjustment, of a user when using predefined measurement protocols with predefined operating parameters. This means that it is proposed to log selection and adaptation activities of users in the case of common image recording devices using measurement recordings that can be adapted by the user, in particular also taking into account the quality of the image recording result, in order to find optimization potential in these measurement records and a clear advantage compared to the automatic selection of suitable measurement protocols. In this case, the input data are also expediently determined and logged, in particular the patient model, so that the input data determined in the training data assigned output data, namely the user settings, are assigned.

As already mentioned, it is a particularly expedient feature of the present invention to use a patient model which, based on the present registration and the adaptation to the current patient, describes as precisely as possible which portion of the patient is placed in the field of view of the image recording device. In this context, a specific embodiment provides that at least the patient model adapted to the current patient is used as the patient data describing the patient, which describes the patient's composition in a locally resolved manner. A patient tomography-based patient model and / or a vector-based patient model and / or a patient model depicting the position of at least some of the organs and / or the course of the patient's blood vessels can expediently be used as the patient model. This means that, for example, a patient tomography-based or computer tomography-like patient model is conceivable, which depicts all the important organs and the blood vessel tree of the patient. Such a computed tomography-based patient model can be derived, for example, from a computed tomography image of the current patient himself, but can also be used as a generic model and is particularly suitable when the image recording device is designed as an x-ray device, since it then implicitly describes the depiction of corresponding anatomical structures by the x-ray imaging. However, it is also conceivable to use a vector-based, simple patient model that maps the outline of a generic patient and the position of all important organs, based on the outline, and, preferably by means of the registration, is adapted to the current patient. All in all, a wide variety of configurations of patient models are conceivable, although in preferred configurations the patient model is fundamentally able, together with the registration, at least the location of important organs relative to the Image recording device or its recording arrangement.

The adaptation of the patient model to the current patient, if it is a generic patient model, can take place, for example, via other patient data, for example from the registration and / or from pre-recordings and / or from other patient information, in particular size and / or weight and / or gender. With particular advantage, the registration can also include the adaptation of the generically prescribed patient model to the current patient.

The registration can take place concretely using sensor data of at least one sensor directed at the patient and / or image data of the patient recorded with the image recording device and / or on the basis of patient data retrieved from an information system and / or on the basis of a configuration of a patient couch of the image recording device. In a specific embodiment, a camera, in particular a 3D camera and / or an infrared camera and / or terahertz camera, and / or an ultrasonic sensor can be used as the sensor, wherein other configurations are also possible. Specifically, a large number of methods can therefore be used to bring about registration of the patient with a coordinate system of the image recording device, which in particular expresses itself in a registration of the patient model adapted to the current patient with this coordinate system.

It is initially conceivable to carry out a camera-based, optical registration, the camera depicting the patient lying on the patient couch and creating a correlation of the patient outline with respect to the patient couch, which in particular also the position of the patient ("head first", "feet first.""," Supine "," Prone ", ...) describes. The use of an infrared camera has the The advantage of being able to see through a possible cover of the patient in a sterile manner. Ultrasound-based detection of the patient's position in relation to the patient bed is also conceivable. Within the framework of the registration, patient data obtained from an information system, for example a hospital information system (HIS) and / or a radiology information system (RIS), and / or patient data entered at the image recording device can be used, which can directly describe, such as the position of the patient is on the patient couch or can also describe the patient more precisely, for example by size, age, gender, weight and the like, such patient data preferably being used to adapt a generically prescribed patient model to the current patient. In addition or as an alternative, an image-based registration is also conceivable, for example the use of a first preliminary image and / or the first image data which are recorded by the patient. Information can be extracted from the image data, for example using databases and / or artificial intelligence, and organ identification is also conceivable. Finally, within the scope of the present invention, registration can also take place on the basis of an embodiment of the patient couch, it being possible in a specific embodiment to use a patient table with an integrated pressure-sensitive film which covers the essential support points of the patient, in particular the head, shoulders, buttocks, Detects calves, feet and arms, and correlates with the patient model for registration.

It should be emphasized once again that the adaptation of the used (generically prescribed) patient model to the current patient is preferably carried out on the basis of registration data collected during registration, so that features of the patient contained in the patient model can be localized in the coordinate system of the image recording device. In other words, the patient registration then establishes a relationship between the coordinate systems of the patient model (and thus the patient) and the image recording device, wherein anatomical features described by the patient model can thus be localized with respect to the image recording device, in particular its recording arrangement, in order to draw corresponding conclusions in the determination algorithm.

Within the scope of the present invention, it can be provided that input data describing the image acquisition target are retrieved from an information system and / or that the image acquisition target is derived from other input data, in particular from a portion of the patient positioned within the field of view of the image acquisition device. The image acquisition target (target application) can be determined with particular advantage from entries in a hospital information system (HIS) and / or radiology information system (RIS), after corresponding information is usually obtained when the information system is registered with the patient. However, it is also particularly advantageous to draw appropriate conclusions from the portion of the patient positioned in the field of view of the image recording device, which is possible on the basis of the registration of the patient model, as described in detail. The portion positioned in the field of view of the image recording device can also be used for plausibility checking or analysis.

In a particularly advantageous embodiment of the present invention, it can be provided that the input data also include at least one instrument data describing the medical instrument used for the patient in the image recording process. Instruments used ("devices") can not only have an influence on the imaging of the type of imaging of the image recording device itself, but in particular also impose certain requirements on the image recording due to its imaging properties, especially when it comes to imaging the medical instrument itself, for example when tracking a catheter, when checking the fit of an implant and like. Corresponding input data describing the medical instrument are therefore useful information when determining the operating parameters by means of the determination algorithm.

Specifically, the instrument data can be determined using an identification device reading an information carrier on the instrument and / or a packaging of the instrument and / or instrument data of an information system and / or by evaluating image data of the image recording device showing the instrument and / or at least one for the Imaging type of the image recording device include relevant property. For example, provision can be made for registering the medical instruments on the image recording device, so to speak, by reading out an information carrier of the medical instrument, for example a bar code and / or an RFID chip, using a suitable recognition device. Additionally or alternatively, information from an information system and / or an image-based comparison can also be queried, the medical instrument, for example, being able to be detected within image data of the image recording device by segmentation algorithms, which in particular also enables properties of the medical instrument to be determined. Accordingly, it is also particularly advantageous if properties are stored for the medical instruments, in particular those that are relevant to the type of imaging of the image recording device. For example, properties of the medical instruments can include at least one material of the medical instrument (for example platinum, plastic, iron, tantalum, iodine, etc.), the geometric extent in at least one dimension (for example 1.2 French catheters, i.e. a catheter with a diameter of 0 , 4 mm, stent diameter, diameter of a stent strut, etc.) can be used to adapt the operating parameters of the image recording device based on this.

If, in one example, a small catheter is to be used, it may be expedient to adapt the operating parameters with regard to a better resolution, in addition to any existing optimization of operating parameters with regard to the area of use of the catheter, so that, for example, different image recording parameters and / or image evaluation parameters are used. if the medical instrument is located in the patient's liver area or in the patient's heart area. Further information about the target area of the medical instrument can also be relevant, for example embolization in the target area and the like.

In the context of the present invention, the positions and / or orientations of movable components of the image recording device, in particular of movable components of the recording arrangement and / or the patient couch, are expediently known, and consequently also their positions and / or orientations (poses) in the coordinate system of the image recording device. Component information of this type, which can also be used as input data and / or is otherwise included in the determination algorithm, can be obtained in different ways, so that, for example, the presence and / or positions / orientations of components of the image recording device from the position of actuators assigned to them and / or sensors assigned to the components are determined. In contrast to the prior art, other sensors can of course also be added here, in particular for components that are placed and / or set manually. In this case, for example, aids, in particular filters and / or collimators and / or rasters, can be treated similarly to medical instruments, in particular also correspondingly detected in (first) image data and / or registered by use of information carriers.

The input data are expediently updated cyclically and / or in the event of changes, at least one adaptation one of the operating parameters is based on the changed input data. This means that the input data described can, if necessary, be updated synchronously or at least from time to time. For example, the angulation of a C-arm and / or generally the position of a patient couch can change during an image sequence in a recording process, so that the recording conditions and / or the examination area change. A specific example of this includes the insertion of a catheter or guidewire at the access (loin or radial), the catheter or guidewire being pushed over the aorta (or radial) to the aortic arch while the patient bed is being moved, so that the tip of the medical Instruments can always be seen in the recorded image data record, covering various organs / examination areas. It is particularly advantageous here to always determine the suitable operating parameters dynamically by means of the determination algorithm and thus to provide improved image quality for all positions of the medical instrument by always keeping the input data and the resulting operating parameters up to date.

In an expedient development, the determination algorithm can determine at least one property of the recording area of the patient that is relevant for the type of imaging used by the image recording device and can be used to adapt at least one operating parameter, in particular an image evaluation parameter.

This will be explained in more detail using a specific example of an X-ray device as an image recording device. For example, the determination of the operating parameters of an optimization of the X-ray spectrum for the image acquisition target can include. Today's dose control and spectral adjustment are based on a generally used mixed model of the essential patient material properties, whereby, for example, a certain composition assumption is made. With the help of the available input data, in particular the determination of the proportion of the patient in the field of view of the image recording device, the anatomy corresponding to the proportion being additionally known from the patient model, a more realistic composition of the patient in the proportion within the range of vision can now be set up in a much more dedicated manner, so that the primary and scattered radiation components expected by suitable components X-ray spectrum selection (tube voltage, pre-filtering, etc.) can be optimized.

It should be noted here that the input data, in particular the patient model, can also describe any existing implants of the patient, which can also be taken into account when selecting or determining suitable operating parameters. For example, information about known implants (knees, hips, etc.) can be used to optimize the X-ray spectrum and the X-ray dose, as can information about collimator inserts and / or filters used.

In particular, filters and / or collimators and / or their position and / or their setting can thus be determined as input data or operating parameters. It should be pointed out that information about filters / collimators as well as about implants can also be determined by image analysis, for example of pre-images or first recorded image data of the patient.

In addition, other physical operating parameters, ie image recording parameters, can also be adapted for optimization based on the knowledge of the proportion of the patient in the field of vision, in particular with regard to the OOI (Organ of Interest). For example, based on the input data and the registration, an angiography of the right coronary artery or the left coronary artery can be predicted with a high probability as the image acquisition target and an optimized choice of the pulse length as an operating parameter can be made automatically. Because in the example of the coronary arteries it is known that the left and right coronary arteries are clear show different movement speeds. The right coronary artery moves much faster than the left coronary artery. If the information as to which coronary artery is to be examined is known (follows from registration, patient model, field of view of the image recording device) and, if appropriate, the image recording target (target application - cardangiography), the pulse length of the X-ray pulses and / or the image frequency can be adapted accordingly to the operating parameters, in order to optimize the temporal resolution and to avoid or reduce movement artifacts.

Another useful area of application for the procedure described is a plurality of individual image recordings which follow one another in time and extensive image recording operations. For the individual image recording times (based on the registration and the patient model), the recording geometry with respect to the patient describing physical operating parameters can be determined, for example system positions comprising the position of a patient couch and / or a recording arrangement (in each case including orientation, if applicable). The system positions are approached one after the other and the images are taken, in particular after an approval by an operator. Such is common, for example, in coronary diagnostics in X-ray devices, in particular with a C-arm. The system positions can then include the position of a patient bed and the angulation of a C-arm. Just as other operating parameters can be determined by an artificial intelligence algorithm, this also applies to such sequences of system positions (or, in general, recording geometries). Training data can be obtained, for example, from similar examinations of other patients.

However, the optimizing determination of operating parameters is not limited to physical operating parameters (image acquisition parameters), but can also relate to image evaluation parameters (image processing parameters). For example, depending on the physical operating parameters selected by the determination algorithm and / or on information about the medical instruments and / or implant information used, the image processing can be optimized image-synchronously, for example what the strength of the temporal and image-based noise suppression, the contrast and dynamic management, the spatial frequency-dependent image sharpness and the like.

In a further, particularly advantageous embodiment of the present invention, it can be provided that the operating parameters describe the automatic output of at least one workflow information, in particular with regard to an aid to be used in the image acquisition and / or to minimize a negative impact on the patient. This means that operating parameters can also relate to the interaction with a user, for example by supporting the user in carrying out a workflow associated with the image acquisition process. For example, he can receive workflow information via appropriate output means, in particular the type and / or positioning of an aid to be used during the image acquisition and / or information about the optimal treatment of the patient during the image acquisition process.

For example, it can be recognized from the input data describing the patient to be examined whether it is a pediatric patient, for example a small child, or an adult. From this, for example, a workflow notice can follow that use anti-scatter grids or remove them from the beam path in pediatric examinations if the image recording device is designed as an X-ray device. It is also extremely useful for X-ray devices to automatically generate workflow information on the imaging device side which relates to an optimization of the local dose load, in particular the skin dose. Here, for example, a change in recording geometries, for example in workflow instructions complementary angulations are proposed in order to minimize the locally applied skin dose for the patient in the case of longer image acquisition processes. In the case of other image recording devices, for example a magnetic resonance device as an image recording device, a workflow notification relating to this can relate to the local and / or global SAR, for example.

In addition to the method, the invention also relates to an image recording device, in particular an x-ray device, having a recording arrangement and a control device designed to carry out a method according to the invention. All statements relating to the method according to the invention can be transferred analogously to the image recording device according to the invention, with which the advantages already mentioned can also be obtained. The control device can comprise at least one processor and / or at least one storage means. To carry out the method according to the invention, the control device can in particular have a determination unit that implements the determination algorithm. Furthermore, a registration unit for registering the patient, in particular by means of the patient model, with a coordinate system of the image recording device can also be present. The image recording device can be controlled by implementing the operating parameters by means of a corresponding control unit.

A computer program according to the invention can, for example, be loaded directly into a memory of a control device of an image recording device and has program means to carry out the steps of a method according to the invention when the computer program is executed in the control device of the image recording device. The computer program can be stored on an electronically readable data carrier according to the invention, which consequently comprises electronically readable control information stored thereon, which comprise at least one computer program mentioned and is designed such that when the data carrier is used it is stored in a Control device of an image recording device perform an inventive method. The data carrier can be a non-transient data carrier, in particular a CD-ROM.

Fig. 1

einen schematischen Ablaufplan des erfindungsgemäßen Verfahrens, und

Fig. 2

eine erfindungsgemäße Bildaufnahmeeinrichtung in Form einer Röntgeneinrichtung.

Further advantages and details of the present invention result from the exemplary embodiments described below and from the drawings. Show:

Fig. 1

a schematic flow chart of the method according to the invention, and

Fig. 2

an image recording device according to the invention in the form of an X-ray device.

In the following, exemplary embodiments of the present invention are discussed with reference to an X-ray device as an image recording device, an application of course also being conceivable to other image recording devices of higher complexity, for example magnetic resonance devices.

The basic idea is based on part or all of the input information available, such as a patient model adapted to the current patient, registration of the patient model and thus the patient with a coordinate system of the image recording device, knowledge of the image recording target, information on the current examination area (Organ of Interests - OOI ), Information on implants, direct radiation, position of filters / collimators, inclusion of expected movements in the examination area, knowledge of medical instruments used or the most critical medical instrument for imaging, to select the most optimal possible physical and image processing operating parameters by a control device of the image recording device. For this purpose, an enabled detection algorithm is used, which prefers the complete image-relevant parameterization of the image recording device undertakes so that no user action, for example the selection of a measurement protocol / organ program, is necessary.

This basis of the procedure according to the invention is determined with the aid of the flow chart of Fig. 1 explained in more detail. The input information for the determination algorithm, which can also include at least one artificial intelligence algorithm, in the present case include various input data 2 to 5 in addition to the already mentioned registration 1 of the patient, the patient model (input data 2) first being explained in more detail.

In the exemplary embodiment described here, a fundamentally generic patient model is used, which is adapted to the current patient on the basis of the registration 1 and is registered with the coordinate system of the image recording device. A patient tomography-like patient model, which depicts all important organs and the blood vessel system, and / or a vector-based simple patient model, which depicts the outlines of a generic patient and the position of all important organs in relation to the outlines, can be used as the patient model. The registration data received in the course of the registration not only describes the position and the position of the patient, but also his other relevant properties, in particular his dimensions / outlines, so that the patient model is adapted to the current patient and, based on the registration, also a suitable one Positioning of the patient model to the coordinate system of the image recording device is easily algorithmically possible on the control device of the image recording device. The combination of patient model and registration described in this way represents one of the essential foundations of the procedure described here.

Various procedures, in particular also combined, can be used to produce the registration 1 For example, a camera-based optical system that images the patient on the patient couch of the image recording device and creates a correlation of the patient outline with respect to the patient couch, an infrared-based system that works similarly to the camera measuring in the visible range, but covers similar to a terahertz camera can measure, an ultrasound-based detection system with corresponding ultrasound sensors, the use of a pressure-sensitive film integrated into the patient bed, which can measure the essential support points and can correlate with the patient model, as well as additional information about the patient, for example inputs regarding the position of the patient on the patient bed, and patient-specific patient data, in particular height, age, gender, weight and the like. Patient data of this type can be called up, for example, from information systems, in particular from a hospital information system (HIS) and / or a radiology information system (RIS).

Further input data 3 describe the system state of the image recording device, in particular the current recording geometry defined by the recording arrangement. In the case of the x-ray device, for example, the positioning of the x-ray emitter and the x-ray detector as a recording arrangement, which can be carried, for example, by a C-arm, can be described in relation to the patient bed, which is extremely important information in the context of the exemplary embodiments described here, because in conjunction with the The patient model present in the coordinate system of the image recording device (input data 2, registration 1), which describes relevant organs / anatomical structures of the patient, can immediately determine which portion of the patient is in the field of view of the image recording device, which is a clear indication of the image recording goal or the Organ of Interest (OOI). Further input data 5, which relate to the image recording process or specifically the image recording target, can for example, can be retrieved from one of the information systems already mentioned. A corresponding combination of the knowledge of the proportion of the patient in the field of vision (registration 1, input data 2, 3) with the further input data 5, which relate to the image acquisition target, results in particularly precise knowledge of the operating parameters that are likely to be required or their values. If, for example, the input data 5 indicate that cardangiography is to be carried out and the portion of the patient located in the field of view of the image recording device indicates that the right coronary artery is to be recorded, the background knowledge that the right coronary artery is moving faster than the left Coronary artery can be used, for example, to select correspondingly short pulses or a corresponding frame rate, to minimize movement artifacts and to optimize the temporal resolution.

With regard to the input data 3 (operating state of the image recording device), it should also be noted that corresponding information is often available anyway as feedback from actuators and / or by appropriate sensors in the control device of the image recording device or can be easily supplemented. In addition, the input data 3 can also include information about the aids / optional and / or adjustable components used, in particular with regard to collimators and / or filters. Such input data 3 can also be derived from image data of the image recording device itself.

In addition, information about implants of the patient, which can be determined as belonging to the input data 2, can also be derived from image data or retrieved from an information system. Particularly in the case of x-ray imaging as a type of imaging, implants can have an influence on account of their weakening properties, so that operating parameters can be adjusted to the presence of the same.

Another extremely useful input information is represented by the input data 4, which relates in particular to medical instruments ("devices") used on and / or in the patient during the image recording process. Such medical instruments can also be related to the image acquisition target and / or influence the image acquisition, which is why knowledge of them can be extremely useful. Information on the devices used can be determined by reading out information carriers, for example bar codes, by means of a corresponding recognition device, wherein an image-based comparison of the instrument properties can optionally take place, possibly with the aid of segmentation algorithms. The input data 4 also include properties stored for the medical instruments, such as the material, the geometric extent in at least one dimension, and the like, which can also be used in the determination of the operating parameters.

The determination algorithm 6 is ultimately carried out in two steps S1 and S2. In step S1, physical operating parameters, which can also be referred to as image acquisition parameters and relate to the physical image acquisition process, are first determined. Depending on their dependency, operational processing parameters (image evaluation parameters) are then determined in a step S2. As already mentioned, the determination algorithm 6 can comprise at least one artificial intelligence algorithm that is trained, for example, by training data that can be determined by logging when using image recording devices that work based on the measurement protocol. Artificial intelligence lends itself particularly well to more complex sub-issues, while many of the issues described, as indicated by the description, can already be developed without the use of artificial intelligence.

In specific exemplary embodiments, it is possible, for example, to determine the operating parameters describing the x-ray dose and the x-ray spectrum as image acquisition parameters by determining composition information of the irradiated area based on the known proportion of the patient in the field of vision (and possibly knowledge about implants), on the basis of which the corresponding operating parameters can be chosen, whereby artificial intelligence approaches can also be used in this context. An optimization can take place, for example, with regard to the expected primary and scattered radiation components. Knowledge about implants can also influence the display of collimators and / or the use of filters to optimize the X-ray spectrum. With regard to the pulse length and / or the image recording frequency as operating parameters, information on the movement of the organ of interest (OOI) can be determined and evaluated accordingly, for example. Depending on the selected physical operating parameters and the input data 2 to 5, image evaluation parameters are then determined as operating parameters in step S2, for example the strength of the temporal and image-based noise suppression, the parameters describing contrast and dynamic management, the image frequency-dependent image sharpness and the like.

At this point it should also be noted that as a further class of operating parameters, for example in a further step of the determination algorithm 6, which is not shown here for the sake of clarity, operating parameters relating to workflow information to be output can also be determined. Such workflow notices, which are generated automatically, can point out to the user components or aids to be used or not used, for example a scattered radiation grid, and / or can reduce the skin dose to the user by means of corresponding workflow notices or the like can be achieved.

In a step S3, the determined operating parameters are then used to control the image recording device, in particular thus for image acquisition, image processing, image display and image storage.

As the arrow 7 indicates, the input data described can be updated synchronously, for example if the recording geometry and / or the position of the patient bed changes during an image sequence. The reason for this can be, for example, the movement of a medical instrument that is to be tracked within the patient. In this way, operating parameters that are optimally selected based on the current circumstances can always be determined by means of the determination algorithm 6, which is then also carried out corresponding to the image.

It should be noted that the operating parameters can also automatically include positions of the patient couch (and thus (based on the registration) the patient) and / or the recording arrangement, summarized system positions, and their number for successive image recordings with the image recording device. For example, the input data 2 to 5 can describe that a coronary examination should take place, in particular with special imaging requirements, but also describe the position, orientation and size of the heart based on the patient model. From this, suitable recording geometries can be determined for examinations comprising several image recording processes and described as operating parameters, for example patient bed positions and recording arrangement positions, that is to say certain system positions, which can be approached automatically.

The system positions can be machine-learned in everyday clinical practice, for example, from training data describing similar examinations. Here, as in other specific applications, it is possible to, for example, clinic-specific and / or user-specific learn and thus also use the locally and / or individually preferred operating parameters.

Fig. 2 finally shows a representation of an image recording device 8 designed as an interventional x-ray device, which in the present case comprises a C-arm 10 mounted on a movable robot arm 9, on which an x-ray emitter 11 and an x-ray detector 12 are arranged opposite one another as a recording arrangement. This makes it possible to implement different receiving geometries with respect to a patient couch 13, which can also be designed to be movable, on which the patient 14 to be examined and / or treated can be supported. The operation of the image recording device 8 is controlled by a control device 15, which is also designed to carry out the method according to the invention.

The control device 15 can be assigned an output means 16 for outputting workflow information, for example a screen and / or an acoustic output means. In order to execute the determination algorithm 6, the control device 15 can in particular have a corresponding determination unit 17 in addition to other functional units. The control device 15 can also be designed for communication with external information systems 18.

Although the invention has been illustrated and described in detail by the preferred exemplary embodiment, the invention is not restricted by the disclosed examples and other variations can be derived therefrom by a person skilled in the art without departing from the scope of protection of the invention.

Claims (17)

The method wherein the control of the image pickup device (8) based on (8) to be reacted operating parameters takes place for operation of a medical imaging device (8) as part of an image pickup target serving image pickup operation on a patient (14) of by a control device (15) of the image pickup device, characterized characterized in that the operating parameters are determined automatically by means of a determination algorithm (6) from at least the patient (14) in the form of a patient model and / or input data (2, 3, 4, 5) describing the image acquisition target and a registration (1) of the patient (14 ) determined with a coordinate system of the image recording device (8) and used to control the image recording device (8). A method according to claim 1, characterized in that the determination algorithm (6) comprises at least one artificial intelligence algorithm that has been trained by machine learning. Method according to Claim 2, characterized in that training data for the artificial intelligence algorithm are determined by logging an activity, in particular comprising at least one individual parameter adaptation, when using predefined measurement protocols with predefined operating parameters. Method according to one of the preceding claims, characterized in that the operating parameters also include at least one image evaluation parameter in addition to image recording parameters. Method according to one of the preceding claims, characterized in that at least the patient model adapted to the current patient (14) is used as the patient data describing the patient (14), which locally resolved describes the composition of the patient (14). Method according to one of the preceding claims, characterized in that the registration (1) using sensor data of at least one sensor directed at the patient (14) and / or image data of the patient (14) recorded with the image recording device (8) and / or on the basis of patient data obtained from an information system (18) and / or on the basis of a configuration of a patient couch (13) of the image recording device (8). Method according to claim 6, characterized in that the patient model used is adapted to the current patient (14) on the basis of registration data collected during registration (1), so that features contained in the patient model can be localized in the coordinate system of the image recording device (8). Method according to one of the preceding claims, characterized in that input data (2, 3, 4, 5) describing the image recording target are retrieved from an information system (18) and / or derived from other input data (2, 3, 4, 5), in particular from a portion of the patient (14) positioned within the field of view of the image recording device (8). Method according to one of the preceding claims, characterized in that the input data (2, 3, 4, 5) also comprise at least one instrument data describing the medical instrument used for the patient (14) in the image recording process. A method according to claim 9, characterized in that the instrument data using a recognition device reading out an information carrier on the instrument and / or a packaging of the instrument and / or Instrument data of an information system (18) and / or by evaluating image data of the image recording device (8) showing the instrument and / or comprising at least one property relevant to the imaging type of the image recording device (8). Method according to one of the preceding claims, characterized in that the presence and / or positions of components of the image recording device (8) are determined from the position of actuators assigned to them and / or by sensors assigned to the components. Method according to one of the preceding claims, characterized in that the input data (2, 3, 4, 5) are updated cyclically and / or in the event of changes, an adaptation of at least one of the operating parameters on the basis of the changed input data (2, 3, 4, 5 ) he follows. Method according to one of the preceding claims, characterized in that the determination algorithm (6) determines at least one property of the recording area of the patient (14) which is relevant for the type of imaging used by the image recording device (8) and is used to adapt at least one operating parameter. Method according to one of the preceding claims, characterized in that the operating parameters include the automatic output of at least one workflow information, in particular with regard to an aid to be used in the image acquisition and / or to minimize a negative impact on the patient (14). Image recording device (8), in particular X-ray device, comprising a recording arrangement and a control device (15) designed to carry out a method according to one of the preceding claims. Computer program which carries out the steps of a method according to one of claims 1 to 14 when it is executed on a control device (15) of an image recording device (8). Electronically readable data carrier on which a computer program according to claim 16 is stored.

Images